By Anatoly Liberman
A reader commented on my recent statement that Engl. good and god are unrelated and noted that this statement, in addition to being counterintuitive and undemonstrable, can even lead to schisms. Being a peaceful man, I am very much against all kinds of hostilities. Nor do I think that the history of words should interfere with faith to such an extent as to result in religious wars. But god and good are indeed unrelated, and I decided not to wait for the last Wednesday of November, when my monthly gleanings are due, and to say what is known about the origin of the words in question as early as possible (now people say only to not wait; for me such a split infinitive is worse than a schism).
Good has transparent etymology: gather and -gether are related to it. Their root means “fit, suitable.” This circumstance is borne out by numerous cognates in and outside Germanic. That is “good” which has been “fixed,” “assembled,” “put together” in a proper way. By contrast, the origin of god is debatable, which does not mean that we know nothing about its derivation. But before I come to the point, let me say that already long ago the proximity of good and god (in the other Germanic languages the two words also sound alike) gave rise to the conclusion that such a striking similarity in sound cannot be fortuitous. Here are three quotations dated 1589, 1606, and 1637 respectively. I have borrowed them from the book Folk-Etymology by the Reverend A. Smythe Palmer (1883). His etymologies should be treated with caution (though, naturally, he explains why good and god are unrelated), but his collection of examples is excellent. I have partly modernized the spelling of the originals.
“If that opinion were not [that is, if the opinion that god and good are related proved false], who would acknowledge any God? The very Etimologie of the name with us of the North partes of the world declaring plainely the nature of the attribute, which is all one as if we said good [bonus] or a giver of good things.” (1589) “God is that which sometimes Good we nam’d, / Before our English tongue was shorter fram’d.” (1606) “An indifferent man may judge that our name of the most divine power, God, is…derived from Good, the chiefe attribute of God.” (1637)
It could not escape the readers’ notice that I spelled god with low case g. I did it for a reason. The concept of God, of one Supreme Being, was alien to polytheistic religions. The further back we step into the past, the clearer it becomes that at one time people believed in multitudes of beings controlling our fate. Those invisible spirits were revered, worshipped, or propitiated, if you will, to prevent them from making humans ill. Language has preserved multiple traces of that state of mind. Elves possessed arrows and caused back pain (lumbago): their victims were “elf-shot.” Dwarfs, if my etymology of the word dwarf is correct, made people dizzy (“*dwysig”; the asterisk means that such a form has not been attested; the Old English word was dysig, with *w lost before long y), while trolls seem to have made the inhabitants of the earth “droll” (that is, ridiculous, behaving like buffoons, crazy). The situation with the gods (in the plural!) is especially clear. The Greek for “god” is theos. We find the same root in enthusiastic, or “possessed by a god,” which could mean “deranged” or “divinely inspired.” (Engl. enthusiastic is from French; Greek is its ultimate source.) The Germanic gods made one “giddy” (Old Engl. *gydig—a close parallel to enthusiastic). One can see that the spirits above were not thought of as good. The contrary is true.
With the advent of Christianity, dwarfs, trolls, elves, and the pagan gods, along with witches, giants, revenants, and the rest survived in folktales and superstitions. Even before that they descended from their heights and became anthropomorphic. Originally the singular form god did not exist in the Old Germanic languages; only the plural did. Three grammatical genders were distinguished: masculine, feminine, and neuter. The form of the word for “gods” was neuter plural, the most typical choice for designating such multitudes. Some other modern Indo-European words for “god” are unlike god: compare Greek theos, Latin deus, and Slavic bog. It may be that god does not even have a Germanic etymology. Perhaps the early Germanic-speakers borrowed it from the indigenous population of the lands on which we find them in the historical period. However, since in this case the pre-Indo-European substrate that could have lent god to Germanic is beyond reconstruction (substrate being a technical term for a language submerged in the language of later settlers), reference to it by a language historian is tantamount to an admission of final defeat. Hence the many attempts to find an Indo-European cognate of god. Any “thick” dictionary will inform us that god can be compared with two Sanskrit words: one meaning “to invoke,” the other “to pour.” Today most etymologists prefer the second hypothesis and interpret “pour” as “libation” (in the process of sacrifice), but the idea of invocation also has learned supporters.
My opinion does not weigh more than either of those two, but I believe that both conjectures are wrong. The primitive “gods” may have been invoked or sacrificed to, but the main thing about them was that they were feared. That is why I share the idea of Karl Brugmann, a great German scholar, who was active in the 19th and at the beginning of the 20th century. He also found only a Sanskrit word to guide him, namely the adjective ghoras “awful, frightening.” According to him, Greek theos had the same origin. If he had been right, the result would have been illuminating, but, as it turned out, theos is not related to the Sanskrit adjective, and Brugmann’s etymology lost interest in the eyes of his colleagues. Yet even though theos and god do not belong together, it does not follow that god and ghoras should be kept apart. I think they possibly are, but hardly anyone will side with me. Likewise, I am in the absolute minority in my conviction that Slavic bog “god” is related to such English words as bug, bogy(man), and their kin. The inherent weakness of the etymologies cited above—from “invoke,” “pour,” and “frightening”— is (apart from the uncertainty of our word’s Indo-European provenance) that a single putative cognate of the Germanic word turns up so far from Germanic, in the language of Ancient India. A search for a better solution continues. Not long ago god was represented as the sum of the particle g- “that one” and an old root meaning “upward.” There also are several older etymologies that have been rejected as untenable, because they are untenable. Of the four words—theos, deus, bog, and god—only deus poses no problems: it is related to Zeus’s name and refers to a bright sky; here we are dealing with a primitive sky god.
After the conversion to Christianity, a word for “God” became necessary, and it had to belong to the masculine gender. This is indeed what happened: the singular was abstracted from the plural, and the neuter yielded to the masculine. Whatever the etymology of god may be, god and good are not related. I should also say that reference to intuition, if intuition means an undisciplined emotion, should be avoided. Etymology is a study of word history and presupposes a professional look at the development of sounds, grammatical forms, and meaning in many languages. “Intuitively,” deus and theos are two variants of the same word, but they are not. The term folk etymology covers suggestions of the theos-deus and god-good type: the temptation to connect look-alikes is irrepressible, but, unless we choose to remain in pre-scientific etymology, it should be resisted. Although “scientific etymology” stumbles at every step, there is no need to make it limp even more by burdening it with naïve medieval hypotheses. I sincerely hope that no schism will be the result of this post.
Anatoly Liberman is the author of Word Origins…And How We Know Them as well as An Analytic Dictionary of English Etymology: An Introduction. His column on word origins, The Oxford Etymologist, appears here, each Wednesday. Send your etymology question to firstname.lastname@example.org; he’ll do his best to avoid responding with “origin unknown.”